JP2015130659A - Learning hearing aid - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a hearing aid system including a geographical position and user feedback in determining the category of a sound environment for automatic adjustment of signal processing parameters.SOLUTION: A hearing aid system 10 includes: a single first hearing aid 12 including a posture sensor 44; and a hand-held device 30 including a GPS receiver 48, a sound environment detector 14 and a user interface 45. The sound environment detector 14 records a determined geographical position together with the determined category of the sound environment at each geographical position. The sound environment detector is configured to record the adjustment executed by using the user interface 45, to correct the automatic adjustment of at least one signal processing parameter (θ of Θ) on the basis of Bayesian incremental preference elicitation in accordance with the recorded adjustment and to perform the corrected automatic adjustment when the same sound environment is detected next.

Description

  New hearing aid systems provide improved automatic selection and adjustment of hearing aid signal processing parameters in response to audio environment, geographic location, and user feedback. In particular, the novel hearing aid system features optimization of hearing aid signal processing parameters based on incremental preference derivation based on geographic location and Bayesian estimation.

  Conventional hearing aids today are typically equipped with a digital signal processor (DSP) for processing the sound received by the hearing aid to compensate for the hearing loss of the user. As is well known in the art, DSP processing operations include gain in each frequency channel of a multi-channel hearing aid, corner frequency or slope of the frequency selective filter algorithm, knee point and compression rate of the compressor algorithm. Is controlled by a signal processing algorithm having various parameters for adjusting the actual signal processing operations to be performed, such as parameters controlling the.

  DSP flexibility is often used to provide a number of different algorithms with different signal processing parameters. For example, various algorithms may be provided for noise suppression, i.e., to attenuate unwanted signals and amplify desired signals. The desired signal is usually utterance or music, and the undesired signal can be background utterance, sound that the tableware touches, music (when utterance is the desired signal), traffic noise, etc.

  Various algorithms and parameters are typically comfortable and comprehensible playback in various categories of voice environments such as speech, bubble utterances, restaurant touches, music, traffic noise, etc. Included to provide voice quality.

  Audio signals obtained from different voice environments may have very different characteristics, such as average and maximum sound pressure level (SPL) and / or frequency components. Thus, in a hearing aid with a DSP, each category of audio environment has a specific signal processing parameter setting that provides a processed audio with optimal signal quality for the category of audio environment of interest. It can be associated with a signal processing algorithm.

  As a result, today's DSP-based hearing aids are typically provided with a number of different signal processing algorithms, each algorithm tailored to a particular category of audio environment and / or a particular user preference. Yes. The signal processing parameters are typically determined at the first fitting in the hearing aid dispenser office, invoking the desired algorithm to set the algorithm parameters in the non-volatile memory area of the hearing aid, and / or desired The algorithm and algorithm parameter settings are programmed into the hearing aid by sending them to a non-volatile memory area.

  Some known hearing aids represent the user's voice environment as one of several categories of voice environments, such as speech, bubble utterances, tableware touch sounds, music, traffic noise, etc. It is possible to classify automatically.

  Use the acquired classification results within the hearing aid to automatically select the signal processing characteristics of the hearing aid, for example, automatically switch to the most suitable signal processing algorithm and parameters for the environmental category of interest Can be considered. Such hearing aids could automatically maintain optimal speech quality and / or speech comprehension for individual hearing aid users in various categories of speech environments.

  US Patent Application Publication No. 2007/0140512 (A1) and WO 01/76321 disclose examples of classifier techniques.

  The new hearing aid system is provided with a hearing aid, which includes the geographical location of the new hearing aid system and user feedback in determining the category of the audio environment.

  A voice environment within a certain geographic area typically remains in the same category over time. Thus, incorporating geographic location into the current audio environment category determination would improve the category determination, i.e., the category determination could be made faster and / or the category determination would be It can be done with higher certainty.

A new hearing aid system is provided, the hearing aid system comprising:
(A) a first hearing aid,
A first microphone,
A first microphone for providing a first audio input signal in response to an audio signal received by the first microphone in an audio environment;
A first processor configured to process a first audio input signal according to a signal processing algorithm F (Θ) to generate a first hearing loss compensated audio signal, wherein Θ is a set of signals A first hearing aid having a first processor, a processing parameter, and a first output transducer for providing a first acoustic output signal based on the first hearing loss compensated audio signal;
(B) a location detector configured to determine the geographical location of the hearing aid system;
(C) a first audio environment detector,
Making a determination of the category of the audio environment surrounding the hearing aid system based on the audio signal received by the hearing aid system and the determined geographic location of the hearing aid system; and
Providing a first output to the first processor to select a first value of the set of signal processing parameters Θ based on the category of the speech environment determined by the first speech environment detector. like,
A first audio environment detector configured;
(D) a user interface for allowing a user of the hearing aid system to make an adjustment of at least one signal processing parameter θ∈Θ;
The first audio environment detector is
Recording the adjustment of at least one signal processing parameter θ∈Θ made by a user of the hearing aid system; and
Based on the adjustment, so as to provide the first output to the first processor,
It is configured.

  Providing the first output to the first processor may be based on an incremental preference derivation based on Bayesian estimation for the adjustment.

  The hearing aid system has a library of signal processing algorithms F (Θ) that includes parameters that control the choice of algorithm to execute, where Θ is the algorithm parameter space, for example, noise suppression algorithms are noisy environments It can be considered that it can be selected for execution and not selected for execution in a quiet environment.

  The location detector is a GPS receiver, a calendar system, a WIFI network interface, and a mobile phone network interface to determine the geographic location of the hearing aid system and possibly the speed of the hearing aid system. Including at least one.

  The first audio environment detector includes a signal of the audio signal received by the hearing aid system, the determined geographical location of the hearing aid system, the date, time, the speed of the hearing aid system, and the signal received by the GPS receiver. The audio environment category surrounding the hearing aid system may be determined based on at least one parameter selected from the group consisting of intensity.

  The voice environment at a particular geographic location, such as a city block, may be due to repeated fluctuations in traffic, number of people, etc., for example, repeating within a year, in a similar manner every year, and / or It is conceivable to change within a day and in a similar manner every day, and such fluctuations may occur when the voice environment detector determines the date and / or time when determining the category of the voice environment. Can be taken into account by allowing

  The signal strength of the signal received by the GPS receiver is significantly reduced when the hearing aid system is inside the building, and therefore information about the GPS signal strength is whether the hearing aid system is inside the building. Can be used by the audio environment detector to determine

  For example, information about the speed of movement determined by the GPS receiver can be used by the audio environment detector to determine that the hearing aid system is inside a transportation device, such as an automobile.

  Hearing aids can be any type configured to be head-mounted on the head and shift position and posture with the head, such as BTE, RIE, ITE, ITC, CIC, and other hearing aids Can be.

  Throughout this disclosure, the term GPS receiver is a satellite maintained by the US government and freely accessible to anyone using a GPS receiver, typically referred to as a “GPS system” Positioning system, Russian Global Positioning Satellite System (GLONASS), Galileo Positioning System of the European Union, Compass Positioning System of the People's Republic of China Satellite system (Regional Navigational 20 Satellite System), etc., and StarFire, Omnistar, India G PS assisted geostationary satellite orbit augmentation positioning (GAGAN), European geostationary satellite positioning overlay service (EGNOS), and Japan multipurpose satellite augmentation system (MSAsurit: ulf It is used to refer to a satellite signal receiver for any satellite positioning system that provides location and time information anywhere on or near the Earth, including augmented GPS, such as Augmentation System). In augmented GPS, the ground-based reference station network measures small fluctuations in the GPS satellite signal, a correction message is sent to the GPS system satellite, and the GPS system satellite broadcasts the correction message to the earth. The enhanced GPS-enabled receiver uses the correction while calculating their position to improve accuracy. The International Civil Aviation Organization (ICAO) refers to this type of system as a satellite-based augmentation system (SBAS).

  The hearing aid is further configured to determine one or more of the posture of the head of the user wearing the hearing aid, eg, head yaw, head pitch, head roll, or combinations thereof, eg, tilt or tilt. One or more attitude sensors, such as gyroscopes, eg, MEMS gyros, tilt sensors, roll ball switches, etc., configured to output signals.

  Throughout this disclosure, a calendar system is a system that provides users with an electronic version of a calendar having data that can be accessed via a network such as the Internet. Well known calendar systems include, for example, Mozilla® Sunbird®, Windows® Live Calendar, and Google® Calendar. (Google Calendar), Microsoft (registered trademark) Outlook (registered trademark) and Exchange Server (Microsoft Outlook with Exchange Server).

  Throughout this disclosure, the term “tilt” refers to the angular deviation from the normal vertical position of the head when the user is standing or seated. Accordingly, the inclination is 0 ° at the stationary position of the head of a person standing or sitting, and the inclination is 90 ° at the stationary position of the head of a person lying on the side.

  A first audio environment detector for selecting a first value of a set of signal processing parameters Θ based on a user's head posture determined based on output signals of one or more posture sensors; And providing a first output. For example, if the user changes posture from sitting to lying down to take a nap, the environment detector may cause the first signal processor to switch the signal processing algorithm accordingly, for example, the first hearing aid Can be automatically muted.

  Alternatively, the output signal of the one or more attitude sensors is a first of a set of signal processing parameters Θ based on the output signal of the one or more attitude sensors and the output of the first audio environment detector. It can be input to another part of the hearing aid system, eg, the first processor, configured to make a value selection.

  The signal processing algorithm may comprise multiple sub-algorithms or sub-routines, each performing a specific subtask within the signal processing algorithm. As an example, the signal processing algorithm may comprise different signal processing sub-routines such as frequency selective filtering, single channel or multi channel compression, adaptive feedback suppression, speech detection, noise reduction, etc.

  In addition, several distinct choices of signal processing algorithms, sub-algorithms, or sub-routines can be grouped together and the user can choose between them according to his / her own preferences Two, three, four, five, or more preset different listening programs can be formed.

  A signal processing algorithm has one or several associated algorithm parameters. These algorithm parameters may typically be divided into smaller parameter sets, each such algorithm parameter set being assigned to a particular part of the signal processing algorithm or to a particular sub-set. • Related to routines. These parameter sets include their respective algorithms or sub-bands, such as filter corner frequency and slope, compressor algorithm compression threshold and compression ratio, fitness feedback suppression algorithm precision and probe signal characteristics, etc. • Control certain characteristics of the routine.

  The value of the algorithm parameter is preferably stored intermediately in a volatile data memory area of the processing means, such as a data RAM area, during execution of the respective signal processing algorithm or sub-routine. The initial values of the algorithm parameters are stored in a non-volatile memory area, such as an EEPROM / flash memory area or a battery back-up RAM memory area, and the user removes or replaces the hearing aid battery, or It enables these algorithm parameters to be retained during the power down that normally occurs by operation of the OFF switch.

  For example, location detectors, including GPS receivers, determine the user's geographic location based on satellite signals in a well-known manner when the user wears a hearing aid on the head in the intended operating position. To be included in the first hearing aid. Thereby, the current position of the user and possibly the orientation can be provided, for example, to the first audio environment detector based on the data from the first hearing aid.

  A voice environment detector may store hearing aid parameters with GPS data on a remote server, eg, a remote server accessed via the Internet, and in some cases, for example, at various GPS locations. It may be configured to be stored with the user's hearing profile for backup of hearing aid settings and / or for sharing of hearing aid settings at various GPS locations with other hearing aid users.

  Thus, the voice environment detector may be configured to retrieve another user's hearing aid settings made at the current GPS location. Hearing aid settings may be grouped according to similarities in hearing profiles, and / or age, and / or race, and / or ear size, etc. Can be selected according to belonging to.

  The first audio environment detector may be included in the first hearing aid, thereby facilitating signal transmission between the audio environment detector and other circuit elements of the hearing aid.

  Alternatively, a location detector including, for example, a GPS receiver can be included in a handheld device interconnected with a hearing aid.

  The handheld device is a smart phone having a GPS receiver and, for example, a GPS receiver, such as an iPhone, an Android phone, a Windows phone. (Windows phone), and a calendar system interconnected with a hearing aid.

  The first audio environment detector may be included in the handheld device. The first audio environment detector benefits from the higher computational resources and power typically available in handheld devices compared to the limited computational resources and power available in hearing aids. It can be received.

  The handheld device may contain a user interface configured for user control of a hearing aid system including, for example, a first hearing aid.

  The handheld device may have an interface for connecting to a wide area network (Wide-Area-Network) such as the Internet.

  Handheld devices may access a wide area network via a cellular network such as GSM®, IS-95, UMTS, CDMA-2000, etc.

  Via a wide area network, such as the Internet, the handheld device communicates and to the time management and communication electronic tools used by the user to store time management and communication information associated with the user. You may have access to Tools and stored information typically reside on remote servers that are accessed via a wide area network.

  The hearing aid may comprise a data interface for transmitting control signals from the handheld device to other parts of the hearing aid system, including the first hearing aid.

  The hearing aid may comprise a data interface for transmitting the output of one or more attitude sensors to the handheld device.

  The data interface may be a wired communication interface, eg, a USB interface, or a wireless communication interface, such as a Bluetooth (Bluetooth) interface, eg, a low energy Bluetooth interface. Good.

  The hearing aid may comprise an audio interface for receiving audio signals from handheld devices and possibly other audio signal sources.

  The audio interface can be a wired communication interface or a wireless communication interface. The data interface and audio interface can be combined into a single interface, eg, a USB interface, a Bluetooth interface, etc.

  Hearing aids, for example, a low energy Bluetooth data interface for exchanging sensor and control signals between the hearing aid and the handheld device, and wireless communication for exchanging audio signals between the hearing aid and the handheld device And an audio interface.

  The first audio environment detector may comprise a first feature extractor for determining a characteristic parameter of the first audio input signal.

  The feature extractor may determine characteristic parameters of the audio input signal, such as average and maximum sound pressure level (SPL), signal power, spectral data, and other well known features. The spectral data can include Discrete Fourier Transform coefficients, Linear Predictive Coding parameters, cepstrum parameters, or corresponding differential cepstrum parameters.

  The feature extractor may output the characteristic parameter to a first environment classifier configured to determine a category of the speech environment based on the determined characteristic parameter and geographic location.

  The first environment classifier determines a category of the voice environment in a plurality of voice environment classes or categories such as speech, bubble utterance, sound of tableware touching at a restaurant, music, traffic noise, and the like. It is configured. The classification process may utilize a simple nearest neighbor search, a neural network, a Hidden Markov Model system, or other system capable of pattern recognition. The output of the environmental classification can be a “hard” classification that encompasses only one environmental category, or a set of probabilities that indicate the probability of the voice environment belonging to each category. Other outputs may also be applicable.

  A first environment classifier for a first parameter map configured to provide an output for selection of a first signal processing algorithm and parameters suitable for execution by the first processor; The determined category of the audio environment may be output.

  In this method, the acquired classification results are used in the hearing aid to automatically select the signal processing characteristics of the hearing aid, for example, automatically switch to the most suitable algorithm for the audio environment of interest. Can do. Such a hearing aid would be able to maintain optimal speech quality and / or speech comprehension for individual hearing aid users in various categories of speech environments.

  As an example, it may be desirable to switch between an omnidirectional microphone preset program and a directional microphone preset program depending not only on the background noise level but also on the further signal characteristics of this background noise. . In situations where the hearing aid user is talking to another person in the presence of background noise, it would be beneficial to be able to identify and classify the type of background noise. If the noise is traffic noise, an omnidirectional operation may be selected to allow the user to hear the approaching traffic flow clearly regardless of the direction of arrival. On the other hand, if the background noise is categorized as being bubble noise, a directional listening program is selected, accompanied by an improvement in the signal-to-noise ratio (SNR) during conversation. The target speech signal may be audible to the user.

  By applying a hidden Markov model to the analysis and classification of the microphone signal, for example a detailed characterization of the microphone signal can be obtained. Hidden Markov models can model stochastic and non-stationary signals in terms of both short and long time variations.

  The audio environment detector may be configured to record the geographic location determined by the location detector along with the determined category of the audio environment at the geographic location. Recording operations may be performed at regular time intervals and / or at a certain geographic distance between recording operations and / or certain events, such as shifts in categories of audio environments, signal processing programs, etc. , Changes in signal processing operations such as changes in signal processing parameters, and the like.

  When the hearing aid system is located within a threshold distance from the geographical location of the previously recorded audio environment category and / or of the geographical location of the same recorded audio environment category. When located within range, the audio environment detector is that the current audio environment is of the same category as already recorded at or near the current geographical location. It may be configured to increase the probability or to determine that the current audio environment is of an already recorded category of audio environment.

  The first speech environment detector determines a speech environment category in view of a probability of occurrence for a previously recorded speech environment category that is within a distance threshold from the determined geographic location. Can be configured as follows.

  The distance threshold reflects, for example, the uncertainty in determining the geographic location of the location detector, for example, is less than or equal to the uncertainty of the location detector, or a record of the category of geographic location and voice environment Reflecting in advance that it is less than the average distance between movements, or less than the characteristic size of the prominent features in the current geographical location such as athletic field, central station, city hall, theater, etc. Can be determined. The distance threshold may be adapted to the current environment, for example, in areas where the distance between recording operations of different categories of the audio environment is short, for example in urban areas, a relatively short distance threshold results. In a region where the distance between recording operations of different categories in the audio environment is long, for example in an open region, a relatively long distance threshold is generated.

  The user interface of the hearing aid system may be configured to associate a particular category of audio environment with each geographic area.

  In the absence of a useful GPS signal, the location detector can be based on the address of the WIFI network to which the hearing aid system can be connected, or from various GSM transmitters well known in the art of mobile phones. Triangulation based on the received signal may determine the geographical location of the hearing aid system. In addition, the location detector accesses the user's calendar system to obtain information about the user's expected whereabouts, such as conference rooms, offices, canteens, restaurants, homes, etc. Can be configured to include this information. Thus, information from the user's calendar system could be replaced or supplemented with information about the geographic location determined in other ways, for example by a GPS receiver.

  For example, if the location detector detects that the user is in an airplane, the audio environment detector may automatically switch the hearing aid system hearing aid to flight mode, i.e. the hearing aid radio is turned off. The

  Also, when the user is inside a building, such as a high-rise building, it may be impossible to determine the geographic location by the GPS receiver because the GPS signal is not present or very weak. . Information from the calendar system about the user's whereabouts can then be used to provide information about the geographic location, or the information from the calendar system supplements the information about the geographic location. For example, a display of a particular conference room may provide information about which floor of the high-rise building the hearing aid system is located on. Information about height is typically not available from a GPS receiver.

  The location detector performs automatic use of information from the calendar system when the geographic location cannot be determined otherwise, for example, when the GPS receiver cannot provide the geographic location. obtain. If the location detector does not have any information about the geographic location, for example from a GPS receiver and calendar system, the audio environment detector is based on the received audio signal in a conventional manner, It is conceivable to categorize the audio environment or the hearing aid can be set to operate in a mode previously selected by the user, for example when fitting or when the situation occurs.

  The user may not be satisfied with the automatic selection of parameter values and may use the user interface to perform adjustments to signal processing parameters, for example, the user may select the current selection of signal processing algorithms. May be changed to another signal processing algorithm, for example, a user may switch from a directional signal processing algorithm to an omnidirectional signal processing algorithm.

  The audio environment detector is configured to incorporate user adjustment of signal processing parameter values over time.

  The audio environment detector performs an automatic adjustment of at least one signal processing parameter θεΘ in the hearing aid system using a library of signal processing algorithms F (Θ) that includes parameters that control the selection of the algorithm to be performed. Where Θ is the algorithm parameter space, for example, the noise suppression algorithm is selected for execution in a noisy environment and not selected for execution in a quiet environment.

The voice environment detector
To record the adjustments made by the user of the hearing aid system, and
Depending on the recorded adjustment, modify the automatic adjustment of the at least one signal processing parameter θ∈Θ based on an incremental preference derivation (based on Bayesian estimation), so that the same audio environment is detected next time So that a modified automatic adjustment is performed when
It is configured.

  Bayesian estimation involves the collection of evidence that matches or does not match a given hypothesis. Confidence in the hypothesis changes as evidence accumulates. If there is sufficient evidence, the certainty is often very high or very low.

  Bayesian estimation uses a numerical estimate of hypothesis confidence before the evidence is observed, and calculates a numerical estimate of hypothesis confidence after the evidence is observed.

Bayes' theorem adjusts probabilities in the following manner for a given new evidence.
P (H ₀ | E) = P (E | H ₀ ) P (H ₀ ) / P (E)
Where
H ₀ represents a hypothesis, called the null hypothesis, estimated before new evidence E was available,
P (H ₀ ) is called the prior probability of H ₀ and
P (E | H ₀ ) is called the conditional probability that evidence E is observed if hypothesis H ₀ is true. This is also called a likelihood function when expressed as a function for H ₀ when E is given,
P (E) is called E's marginal probability and is the probability that new evidence E is observed under all mutually exclusive hypotheses.

This P (E) can be calculated as the sum of the products of all the probabilities of mutually exclusive hypotheses and the corresponding conditional probabilities, ie, ΣP (E | H _i ) P (H _i ). P (H ₀ | E) is called the posterior probability of H ₀ when E is given.

The factor P (E | H ₀ ) / P (E) represents the effect of evidence on the hypothesis confidence. This factor increases if evidence is expected to be observed when the hypothesis under consideration is true. Multiplying the hypothesis prior probability by this factor results in a high posterior probability of the hypothesis given evidence. Thus, under Bayesian estimation, Bayesian theorem measures how much new evidence should change the confidence in the hypothesis.

  For more information on Bayes' theorem and Bayesian inference, see "Information Theory, Inference, and Learning Algorithms" David J. et al. C. See Mackay, Cambridge University Press, 2003.

  The Bayesian inference approach to probability theory is a consistent rational theory for inference under uncertainty. Such statistical techniques are needed to address these uncertainties because the perceptual feedback from the listener is (partially) unknown and often unreasonable. In the following, the method of Bayesian estimation for the processing operation of the hearing aid, and in particular, an incremental preference derivation method based on Bayesian estimation will be dealt with in more detail.

  The new hearing aid system audio environment detector has the desired adjustment of signal processing parameters associated with the audio environment and the correct, user adjustments that are personal, dynamic, non-linear, and stochastic. It is possible to effectively learn the complex relationship between Therefore, the sound environment detector can be considered as a learning-type sound environment detector.

  The audio environment detector may update at least one signal processing parameter θεΘ each time a user makes an adjustment. Alternatively, the update operation may be performed, for example, according to certain criteria that the user makes a predetermined number of adjustments so that only significant adjustments lead to updates.

  Sometimes during operation of the device, the user may not be satisfied with the quality of the received signal, and therefore uses the user interface to make adjustments to the hearing aid system. The learning goal is to slowly absorb regular patterns into the model parameter θ by the speech environment detector. Ultimately, the processing leads to a reduction in the number of user operations.

  Parameter updates are performed only when knowledge of user preferences is available. While the user interface is not being operated during normal operation of the device, the user may be satisfied with the quality of the transmitted signal, but this is not certain. Eventually, the user may not be wearing the device. However, when the user starts operating the user interface, it is assumed that the user is not convinced at that moment. The beginning of the operational phase of the user interface is expressed as a moment of disagreement. While the user is operating the user interface, the user may still be looking for better adjustments. The next learning moment occurs immediately after the user stops operating the user interface. At this time, it is assumed that the user has found a satisfactory adjustment, and this is called the moment of consent. The disagreement moment and the consent moment identify the circumstances for collecting disapproval teaching data and approval teaching data, respectively.

In the following, an exemplary method for adapting to user preferences is disclosed. This method is based on incremental preference derivation based on Bayesian estimation, but other methods are possible. Assume that the adjustable signal processing parameters at the kth disagreement moment and the consent moment are set to θ _kd and θ _kc , respectively. Also, assume that the output of the environmental speech classifier during the kth user operation phase remains substantially constant at C _k .

Clearly, under environmental conditions C _k , the user prefers θ _kc to θ _kd (represented by the (end user) decision d _k = θ _kc > θ _kd ). The set of all user decisions up to the _kth decision is D _k−1 = {d ₁ , d ₂ ,. . . , D _k−1 }. A Bayesian update scheme is then used to absorb the kth observation. Let the parameter map from class to hearing aid parameters be represented by a stochastic function p (θ | C, ω), where ω represents a parameter for the parameter map 40.

A Bayesian update to the parameter map, based on the kth user action, then
p (ω | C, D _k ) = p (d _k | C, ω) × p (ω | C, D _k−1 ) / p (C)
Given by.

In Chu and ^{Gharamani (Preference Learning with Gaussian Processes,} 22 nd Int'l conf on Machine Learning, 2005), the equation of the Bayesian update the parameter map 40 based on the Gaussian process, it is used in detail.

  The new hearing aid system may be a binaural hearing aid system having two hearing aids, one for the user's right ear and one for the user's left ear.

Accordingly, the novel hearing aid system is a second microphone, the second microphone for providing a second audio input signal in response to an audio signal received at the second microphone;
A second processor configured to process the second audio input signal according to a second signal processing algorithm F (Θ) to generate a second hearing loss compensated audio signal;
A second hearing aid may be provided having a second output transducer for providing a second acoustic output signal based on the second hearing loss compensated audio signal.

  Because the binaural hearing loss is typically different for the two ears, the second hearing aid will compensate for a hearing loss that is different from the hearing loss typically compensated by the first hearing aid. Except for being adjusted, the second hearing aid circuit is preferably identical to the first hearing aid circuit.

  The first audio environment detector may be configured to determine a category of audio environment surrounding the user of the hearing aid system based on the first and second audio input signals and the geographical location of the hearing aid system.

  The first audio environment detector provides a second output to the second processor for selection of a second signal processing algorithm and parameter F (Θ) executed by the second processor. And may be configured to generate a second hearing loss compensated audio signal.

  Alternatively, the second hearing aid is a second speech environment detector, similar to the first speech environment detector, and the first and second audio input signals and the geographical location of the hearing aid system. To determine a category of the speech environment based on the location and to select a second value of the set of signal processing parameters Θ based on the category determined by the second speech environment detector. A second audio environment detector configured to provide a second output to the second processor.

  In a binaural hearing aid system, it is important that the signal processing algorithms of the first and second signal processors are selected in a coordinated manner. Because the characteristics of the audio environment can be significantly different in the user's two ears, it often happens that the unique determination of the category of the audio environment in the user's two ears is different, which means that the sound in the hearing aid , Which can lead to undesired different signal processing operations. Therefore, the signal processing algorithm of the first and second processors may be an audio signal received at the handheld device, an audio signal received at the left ear and an audio signal received at the right ear, or the handheld device. Is preferably selected based on the same signal, such as a combination of the audio signal received at the left ear and the audio signal received at the left ear and the audio signal received at the right ear, or the like.

  Similar to the first audio environment detector, the second audio environment detector may comprise a second feature extractor for determining a characteristic parameter of the second audio input signal.

  The second feature extractor may output the characteristic parameter to the second environment classifier to determine a category of the speech environment based on the determined characteristic parameter and the geographic location.

  The second environment classifier outputs a voice environment category to a second parameter map configured to provide an output for the purpose of selecting a second signal processing algorithm of the second processor. Can do.

  As already mentioned, the method in the new hearing aid system has the ability to absorb user preferences that change over time and / or typical audio environment changes experienced by the user. Personalization of the hearing aid can be performed during normal use of the hearing aid. These advantages are obtained by absorbing the hearing aid user adjustments in the hearing aid processing parameters. Over time, this approach leads to fewer user operations during periods when the user's preferences do not change. Furthermore, these methods are robust against unreasonable user behavior.

  User preferences for signal processing parameters are derived during normal use in a consistent and rational manner and according to the theory of reasoning under uncertainty.

  The new hearing aid system is a complex between the desired adjustment of signal processing parameters and corrective user adjustment that is personal, dynamic, non-linear and / or stochastic. It is possible to learn the relationship.

  The new hearing aid system can distinguish between different user preferences caused by different audio environments. This allows the signal processing parameters to be automatically adjusted according to the user's perception of the best possible parameter settings for the actual audio environment.

  The medium for storing information / data may take many forms, including but not limited to, non-volatile media, volatile media, and transmission media. The non-volatile medium can be, for example, an optical storage device, a magnetic storage device, or other type of storage device. Non-volatile media may be considered an example of non-transitory media. Volatile media includes dynamic memory, such as main memory. Volatile media can be considered another example of non-transitory media. Transmission media includes cables, wires, and optical fibers. Transmission media can also take the form of acoustic or light waves, such as those generated during radio wave and infrared data communications.

  The signal processing operations in the new hearing aid system can be performed by dedicated hardware, can be performed in a signal processor, or can be performed in a combination of dedicated hardware and one or more signal processors.

  As used herein, the terms “processor”, “signal processor”, “controller”, “system”, etc. refer to hardware, a combination of hardware and software, software, or running software. It is intended to refer to any of the CPU related entities.

  For example, “processor”, “signal processor”, “controller”, “system”, etc. are not limited to the following, but include processes, processors, objects, executable files, execution threads, And / or can be a program.

  By way of example, the terms “processor”, “signal processor”, “controller”, “system”, and the like refer to both an application running on a processor and a hardware processor. One or more “processors”, “signal processors”, “controllers”, “systems”, etc., or any combination thereof may be resident in a processing and / or execution thread. One or more “processors”, “signal processors”, “controllers”, “systems”, etc., or any combination thereof, on one hardware processor, possibly with other hardware circuit elements It can be localized in combination and / or distributed between two or more hardware processors, possibly in combination with other hardware circuits.

  A processor (or similar terminology) can also be any component or any combination of components capable of performing signal processing operations. For example, the signal processor can be an ASIC processor, FPGA processor, general purpose processor, microprocessor, circuit component, or integrated circuit.

  The hearing aid system is (a) a first hearing aid, a first microphone, that provides a first audio input signal in response to an audio signal received by the first microphone in an audio environment. And a first processor configured to process a first audio input signal according to a signal processing algorithm F (Θ) to generate a first hearing loss compensated audio signal. A first processor, wherein Θ is a set of signal processing parameters, and a first output transducer for providing a first acoustic output signal based on the first hearing loss compensated audio signal. A first hearing aid, (b) a location detector configured to determine a geographical location of the hearing aid system, and (c) a first audio environment detector, the hearing aid system comprising: To determine a category of the audio environment surrounding the hearing aid system based on the audio signal received by the system and the determined geographical location of the hearing aid system, and determined by the first audio environment detector A first audio environment detection configured to provide a first output to the first processor to select a first value of the set of signal processing parameters Θ based on the determined category (D) a user interface for allowing a user of the hearing aid to make adjustments of at least one signal processing parameter θεΘ, and (e) at least one signal processing performed by the user of the hearing aid A non-transitory medium for recording the adjustment of the parameter θ∈Θ is provided, and the first sound environment detector is responsive to the first to the first processor based on the adjustment. Is configured to provide output.

  In some cases, the adjustment is based on incremental preference derivation based on Bayesian estimation.

  In some cases, the location detector includes a GPS receiver.

  In some cases, the first audio environment detector is received by the audio signal received by the hearing aid system, the determined geographical location of the hearing aid system, the date, time, the speed of the hearing aid system, and the GPS receiver. The audio environment category surrounding the hearing aid system is determined based on at least one parameter selected from the group consisting of the signal strengths of the received signals.

  In some cases, the hearing aid system further includes a non-transitory medium for recording the geographic location determined by the location detector along with the category of the audio environment at the geographic location.

  In some cases, the first audio environment detector takes into account the probability of occurrence for a previously recorded audio environment category that is within a distance threshold from the determined geographic location, and It is configured to determine a category.

  In some cases, the hearing aid system further includes a non-transitory medium for storing specific categories of the audio environment along with their respective geographic ranges.

  In some cases, the location detector automatically accesses the user's calendar system to obtain information about the user's location, otherwise the location detector determines the geographic location of the hearing aid system. When it is not possible to determine the geographical location of the hearing aid system based on information about the user's location.

  In some cases, the first audio environment detector is configured to automatically switch the first hearing aid of the hearing aid system to flight mode when the location detector detects that the user is in the airplane. .

  In some cases, the first hearing aid comprises at least one posture sensor configured to provide information regarding the posture of the user's head when the user wears the first hearing aid in the intended operating position; The first hearing aid is configured to select the first value based on information related to the posture of the user's head.

  In some cases, the location detector is part of the first hearing aid.

  In some cases, the hearing aid system further includes a handheld device communicatively coupled to the first hearing aid and containing the location detector.

  In some cases, the handheld device also houses a first audio environment detector.

  In some cases, the handheld device comprises a user interface.

  In some cases, the first hearing aid contains a first audio environment detector.

  In some cases, the hearing aid system further includes a second microphone for providing a second audio input signal in response to an audio signal received at the second microphone; A second processor configured to process the second audio input signal according to the signal processing algorithm F (Θ) to generate a second hearing loss compensated audio signal; and a second hearing loss compensated A second hearing aid having a second output transducer for providing a second acoustic output signal based on the audio signal, the first audio environment detector comprising: the first and second audio input signals; The audio environment category is determined based on the geographical location of the hearing aid system.

  In some cases, the first audio environment detector is configured to provide a second output for selecting a second value of the set of signal processing parameters.

  In some cases, the second hearing aid is a second sound environment detector that determines a category of the sound environment based on the first and second audio input signals and the geographical location of the hearing aid system. A second output to the second processor to perform and to select a second value of the set of signal processing parameters Θ based on the category determined by the second audio environment detector A second audio environment detector configured to provide:

  The drawings illustrate the design and utility of the embodiments, in which like elements are designated by common reference numerals. These drawings are not necessarily drawn to scale. In order to better appreciate how the advantages and objectives listed above, as well as other advantages and objectives, may be obtained, a more specific embodiment of the embodiments illustrated in the accompanying drawings will be described. Give an explanation. These drawings depict only typical embodiments and therefore should not be considered as limiting their scope.

FIG. 1 is a diagram illustrating a novel hearing aid system having a single hearing aid with a posture sensor and a handheld device with a GPS receiver, a sound environment detector, and a user interface. FIG. 2 is a diagram illustrating a novel hearing aid system having a single hearing aid with a posture sensor and a sound environment detector and a handheld device with a GPS receiver and a user interface. FIG. 3 is a diagram illustrating a novel hearing aid system having two hearing aids with a posture sensor and an audio environment detector and a handheld device with a GPS receiver and a user interface. FIG. 4 is a diagram illustrating a novel hearing aid system having two hearing aids with posture sensors and a handheld device with a voice environment detector, a GPS receiver, and a user interface.

  In the following, various exemplary embodiments will be described with reference to the figures. It should be noted that these figures are not drawn to scale and that elements of similar structure or function are represented by the same reference numerals throughout the figures. It should also be noted that these figures are only intended to facilitate the description of the embodiments. They are not intended as an exhaustive description of the claimed invention or as a limitation on the scope of the claimed invention. In addition, the illustrated embodiments need not have all of the illustrated implementations or advantages. An implementation or advantage described in conjunction with a particular embodiment is not necessarily limited to that embodiment, and implementation in any other embodiment is not so illustrated, or It may be possible even if not explicitly described as such.

  In the following, the novel hearing aid system will now be described more fully with reference to the accompanying drawings in which various types of novel hearing aid systems are shown. The novel hearing aid system may be embodied in different forms, not shown in the accompanying drawings, and should not be construed as limited to the embodiments and examples specified herein.

  Like reference numbers refer to like elements in the drawings.

  FIG. 1 schematically illustrates a novel hearing aid system 10 having a single first hearing aid 12 having a posture sensor 44 and a handheld device 30 having a GPS receiver 48, an audio environment detector 14, and a user interface 45. Illustratively.

  The first hearing aid 12 may be of any type configured to be head-mounted at the head, such as BTE, RIE, ITE, ITC, CIC, and other hearing aids.

  The first hearing aid 12 is a first front microphone 16 and a first rear microphone 18 responsive to audio signals received by the microphones 16, 18 in an audio environment surrounding the user of the hearing aid system 10. A first front microphone 16 and a first rear microphone 18 connected to a respective A / D converter (not shown) for providing the respective digital input signals 20, 22. The digital input signals 20, 22 are input to a hearing loss processor 24 that is configured to process the digital input signals 20, 22 according to a signal processing algorithm F (Θ) to generate a hearing loss compensated output signal 26. . The hearing loss compensated output signal 26 is sent to a D / A converter (not shown) and an output transducer 28 for converting the hearing loss compensated output signal 26 into an acoustic output signal.

  The new hearing aid system 10 further comprises a handheld device 30, for example a smart phone, that houses a voice environment detector 14 for determining the category of voice environment surrounding the user of the hearing aid system 10. The category determination operation is based on the audio signal picked up by the microphone 32 in the handheld device. Based on the category determination, the audio environment detector 14 provides an output 34 to the hearing aid processor 24 for selection of signal processing algorithms and parameters appropriate for the categorized audio environment.

  Accordingly, the hearing aid processor 24 is automatically switched to one or more algorithms that are most suitable for the categorized voice environment, so that optimal voice quality and / or speech comprehension may be achieved in various voice environments. Maintained. The signal processing algorithm of the processor 24 may perform various other signal processing tasks in addition to various forms of noise reduction and dynamic range compression.

  The first audio environment detector 14 benefits from the computational resources and power typically available in the handheld device 30 that are larger than the resources and power available in the first hearing aid 12.

  The sound environment detector 14 includes a feature extractor 36 for determining the characteristic parameter of the sound signal received from the microphone 32. The parameters can relate to signal power, spectral data, and other well-known features.

  The sound environment detector 14 further includes an environment classifier 38 for determining a category of the sound environment based on the determined characteristic parameter output by the feature extractor 36. The environment classifier 38 classifies voices into a plurality of environmental categories such as utterances, bubble utterances, sounds touched by tableware at restaurants, music, traffic noise, and the like. The classification process may utilize a simple nearest neighbor search, a neural network, a hidden Markov model system, or another system capable of pattern recognition. The output of the environment classifier 38 may be a “hard” determination of the categories that encompass only one environmental category, or a set of probabilities that indicate the probabilities of the voice environment belonging to each category. Other outputs may also be applicable.

  The audio environment detector 14 further comprises a parameter map 40 for providing an output 34 for the purpose of selecting signal processing algorithms and parameters from an available library of signal processing algorithms and parameters F (Θ). . The parameter map 40 maps the output of the environment classifier 38 to a set of parameters θεΘ for the hearing aid audio processor 20. Examples of such parameters are the amount of noise reduction, the amount of gain, the amount of HF gain, the algorithm control parameters that control whether the corresponding signal algorithm is selected for execution, the corner frequency and slope of the filter. The compression threshold and compression ratio of the compressor algorithm, the precision of the suitability feedback suppression algorithm and the probe signal characteristics. Other parameters may be included.

  The handheld device 30 includes a location detector 41 having a GPS receiver 42 configured to determine the geographic location of the hearing aid system 10. The illustrated handheld device 30 is a smart phone that also has a mobile interface 48 with a GSM interface for interconnection with a mobile phone network and a WIFI interface 48 well known in the mobile phone art. It is. In the absence of useful GPS signals, the location of the illustrated hearing aid system 10 can be determined as the address of a WIFI network or received from various GSM transmitters well known in the mobile phone art. May be determined by triangulation based on the signal being transmitted.

  The illustrated audio environment detector 14 is configured to record the determined geographic location along with the determined category of the audio environment at each geographic location. Recording operations may be performed at regular time intervals and / or at a certain geographical distance between recording operations and / or certain events, eg, categories of voice environment, and signal processing programs , Changes in signal processing parameters, changes in signal processing operations such as, etc., and others.

  When the hearing aid system 10 is located within an area of a geographic location where a particular category of audio environment is recorded, the audio environment detector has recorded the current audio environment previously recorded in the audio environment. It is configured to increase the probability of being in each category.

  The user interface 45 of the hearing aid system 10 may be configured to assign a particular category of audio environment to a particular geographic area.

  The illustrated audio environment detector 14 accesses the user's calendar system via, for example, the mobile interface 48 to provide information about the user's whereabouts, such as a conference room, office, canteen, restaurant, home, etc. It is also configured to acquire information and include this information in the determination of the category of the audio environment. Information from the user's calendar system could replace or supplement the information about the geographic location determined by the GPS receiver.

  For example, the audio environment detector 14 may automatically switch the hearing aid of the hearing aid system 10 to flight mode when the user is in an airplane, as shown in the user's calendar system, i.e. The hearing aid radio is turned off.

  Also, when the user is inside a building, for example a high-rise building, the GPS signal may not be present or very weak so that the geographic location cannot be determined by the GPS receiver. It is done. Information from the calendar system about the user's whereabouts can then be used to provide information about the geographic location, or the information from the calendar system supplements the information about the geographic location. For example, a display of a particular conference room may provide information about a floor in a high-rise building. Information about height is typically not available from a GPS receiver.

  The voice environment detector 14 may automatically use information from the calendar system when the GPS receiver is unable to provide a geographic location. If no information about the geographic location is available from the GPS receiver and calendar system, the voice environment detector will categorize the voice environment based on the received voice signal in a conventional manner. Or the hearing aid can be set to operate in a mode previously selected by the user, for example, at the time of fitting or when the situation occurs.

  The hearing aid 12 may be one or more of the posture of the head of the user wearing the hearing aid, eg, head yaw, head pitch, head roll, or combinations thereof, eg, tilt, ie A gyroscope, such as a MEMS gyro, a tilt sensor, a roll ball switch, configured to output a signal to determine angular deviation from the normal vertical position of the head when seated, One or more attitude sensors 44, such as others, are provided. For example, the inclination of the head of a person standing or sitting at the rest position is 0 °, and the inclination of the head of a person lying on the rest position is 90 °.

  The first processor 24 is based on the posture of the user's head determined based on the output signal 46 of the one or more posture sensors 44 and the output control signal 34 of the first sound environment detector 14. The processor 24 is configured to select a first signal processing algorithm. For example, if the user changes posture from sitting to lying down to take a nap, the audio environment detector 14 may cause the signal processor 24 to switch programs accordingly, for example, the first hearing aid 12 Can be automatically muted.

  The environmental classifier 38 adds statistical data from audio signals (such as SNR and RMS) plus location data (from GPS) to environmental classes such as bubble, car, cocktail party, and church. Map. The user may not be satisfied with the automatic selection of parameter values and may use the user interface to perform adjustments to signal processing parameters, for example, the user may select the current selection of signal processing algorithms. May be changed to another signal processing algorithm, for example, a user may switch from a directional signal processing algorithm to an omnidirectional signal processing algorithm.

  The audio environment detector is configured to incorporate user adjustment of signal processing parameter values over time.

  The audio environment detector includes a library of signal processing algorithms F (Θ) that includes parameters that control the selection of algorithms for performing an automatic adjustment of at least one signal processing parameter θ∈Θ within the hearing aid system 10. Where Θ is the algorithm parameter space, for example, the noise suppression algorithm is selected for execution in a noisy environment and not selected for execution in a quiet environment.

The environmental sound detector 14
To record adjustments made by the user of the hearing aid system using the user interface 45; and
Depending on the recorded adjustment, modify the automatic adjustment of at least one signal processing parameter θ∈Θ, based on an incremental preference derivation based on Bayesian estimation, so that the next time the same audio environment is detected So that a modified automatic adjustment is carried out,
It is configured.

  Bayesian estimation involves collecting evidence that matches or does not match a given hypothesis. Confidence in the hypothesis changes as evidence accumulates. If there is sufficient evidence, the certainty is often very high or very low.

  The illustrated hearing aid system includes an audio environment detector that operates to adjust the signal processing parameter θεΘ in response to the audio environment surrounding the hearing aid system 10.

The environment classifier 38 takes an input U, which is a vector of relevant features for the speech environment, including, for example, x _t short-term RMS and SNR estimates. U includes the location of GPS. The output of the environment classifier is represented by a discrete class variable C. Exemplary classes include speech, noise, speech in noise, in the car, in a church, at a cocktail party, etc. The environmental class is mapped onto the hearing aid parameter θ via the parameter map 40.

As noted above, sometimes during operation of the device, the user may not satisfied with the quality of the received signal y _t, therefore, to implement the regulation of the hearing aid system with a user interface 45. The learning goal is to slowly absorb regular patterns into the model parameter θ by the speech environment detector. Ultimately, the processing leads to a reduction in the number of user operations.

  Parameter updates are performed only when knowledge of user preferences is available. While the user interface 45 is not being operated during normal operation of the device, the user may be satisfied with the quality of the transmitted signal, but this is not certain. Eventually, the user may not be wearing the device. However, when the user starts operating the user interface, it is assumed that the user is not convinced at that moment. The beginning of the operational phase of the user interface is expressed as a moment of disagreement. While the user is operating the user interface, the user may still be looking for better adjustments. The next learning moment occurs immediately after the user stops operating the user interface 45. At this time, it is assumed that the user has found a satisfactory adjustment, and this is called the moment of consent. The disagreement moment and the consent moment identify the circumstances for collecting disapproval teaching data and approval teaching data, respectively.

The method of matching user preferences in the hearing aid system 10 is based on incremental preference derivation based on Bayesian estimation, but other methods are possible. Assume that the adjustable signal processing parameters at the kth disagreement moment and the consent moment are set to θ _kd and θ _kc , respectively. Also, assume that the output of the environmental speech classifier during the kth user operation phase remains substantially constant at C _k .

Clearly, under environmental conditions C _k , the user prefers θ _kc to θ _kd (represented by the (end user) decision d _k = θ _kc > θ _kd ). A set of all user decisions up to the _k-th decision is D _k−1 = {d ₁ , d ₂ ,. . . , D _k−1 }. Absorb the kth observation using a Bayesian update scheme. Let the parameter map 40 from class to hearing aid parameters be represented by a stochastic function p (θ | C, ω), where ω represents a parameter for the parameter map 40.

A Bayesian update to the parameter map, based on the kth user action, then
p (ω | C, D _k ) = p (d _k | C, ω) × p (ω | C, D _k−1 ) / p (C)
Given by.

In Chu and ^{Gharamani (Preference Learning with Gaussian Processes,} 22 nd Int'l conf on Machine Learning, 2005), the equation of the Bayesian update the parameter map 40 based on the Gaussian process, it is used in detail.

  The new hearing system 10 shown in FIG. 2 is similar to the new hearing aid system of FIG. 1 except that the audio environment detector 14 has been moved from the handheld device 30 of FIG. 1 to the first hearing aid 12 of FIG. And operate in the same manner. In this manner, the microphone output signals 20, 22 connect directly to the audio environment detector 14 so that the audio environment can be categorized based on the signal received by the microphone in the hearing aid without increasing data transmission requirements. Can be done.

  The novel hearing aid system 10 shown in FIG. 3 includes two hearing aids, a first hearing aid 12A for the user's right ear and a second hearing aid 12B for the user's left ear, a GPS receiver 42 and a mobile A binaural hearing aid system having a handheld device 30 with an interface 48.

  Each of the illustrated first hearing aid 12A and second hearing aid 12B includes signal processing within the two hearing aids 12A, 12B, with the respective audio environment detectors 14A, 14B operating in concert, as further described below. Except for providing coordinated selection of algorithms, it is similar to the hearing aid shown in FIG. 2 and operates in a similar manner.

  Each of the first hearing aid 12A and the second hearing aid 12B of the binaural hearing aid system 10 includes a binaural sound environment detector 14A for determining a category of the sound environment surrounding the user of the binaural hearing aid system 10. 14B. The category determination is based on the microphone output signals 20A, 22A, 20B, and 22B. Based on the category determination, binaural audio environment detectors 14A, 14B provide outputs 34A, 34B to respective hearing aid processors 24A, 24B to select a signal processing algorithm appropriate for the audio environment category. To do. Accordingly, the binaural audio environment detectors 14A, 14B determine the category of the audio environment based on the signals from both hearing aids, ie, in a binaural manner, whereby the hearing aid processors 24A, 24B The system automatically switches to the most suitable algorithm for the category of speech environment, so that optimal speech quality and / or speech comprehension is maintained by the binaural hearing aid system 10 in various speech environments. The

  The binaural sound environment detectors 14A and 14B illustrated in FIG. 3 are the same as the sound environment detector 14 illustrated in FIG. 2, but in FIG. 2, the first sound environment detector 14 is a single hearing aid. 3 differs from FIG. 3 in that each binaural sound environment detector 14A, 14B receives input from both hearing aids 12A, 12B. Accordingly, in FIG. 3, a signal is transmitted between the hearing aid 12A and the hearing aid 12B so that the algorithms executed by the signal processors 24A, 24B are selected in concert.

  In FIG. 3, the output of the environment classifier 14A of the first hearing aid 12A is transmitted to the second hearing aid 12B, and the output of the environment classifier 14B of the second hearing aid 12B is transmitted to the first hearing aid 12A. . Since the parameter maps 40A, 40B of the first and second hearing aids 12A, 12B then operate based on the same two inputs to generate control signals 34A, 34B to make a processor algorithm selection Although the parameter mapping units 34A, 34B receive the same input, the algorithm selection in the two hearing aids 12A, 12B is coordinated.

  The transmission data rate is low because only a set of probabilities or logic values for the category of audio environment need be transmitted between the hearing aid 12A and the hearing aid 12B. A fairly long latency can be tolerated. By applying time constants to variables that change according to the output of the parameter mapping, differences that can be caused by latency can be smoothed. As already mentioned, it is important that the signal processing operations in the two hearing aids are coordinated. However, if a transition period of a few seconds is allowed, the hearing aid system can operate with only 3-4 transmissions per second. Thereby, power consumption is kept low.

  The audio environment detectors 14A, 14B incorporate the determined position provided by the handheld unit 30 of the new hearing aid system 10 in the same manner as disclosed above with reference to FIGS.

  In the novel binaural hearing aid system 10 shown in FIG. 4, one of them 34A is connected to the first hearing aid 12A, and the other 34B is connected to the second hearing aid 12B. A single audio environment detection that is similar to and operates in a similar manner to the audio environment detector shown in FIG. 1 except that the audio environment detector 14 provides two control outputs 34A, 34B. By providing the instrument 14, coordinated signal processing operations in the two hearing aids 12A, 12B are obtained. The illustrated voice environment detector 14 is housed in the handheld device 30.

  Each of the hearing aids 12A, 12B is similar to the hearing aid 12 shown in FIG. 1 and operates in the same manner.

  While specific embodiments have been shown and described, it will be understood that the embodiments are not intended to limit the claimed invention and those skilled in the art will recognize the claims. It will be apparent that various changes and modifications can be made without departing from the spirit and scope of such invention. The specification and drawings are, accordingly, to be regarded in an illustrative sense rather than a restrictive sense. The claimed invention is intended to encompass alternatives, modifications and equivalents.

Claims (18)

A hearing aid system,
(A) a first hearing aid,
A first microphone for providing a first audio input signal in response to an audio signal received by the first microphone in an audio environment;
A first processor configured to process the first audio input signal according to a signal processing algorithm F (Θ) to generate a first hearing loss compensated audio signal, wherein Θ is a set of A first hearing aid comprising: a first processor that is a signal processing parameter; and a first output transducer for providing a first acoustic output signal based on the first hearing loss compensated audio signal;
(B) a location detector configured to determine a geographic location of the hearing aid system;
(C) a first audio environment detector,
Making a determination of the category of the audio environment surrounding the hearing aid system based on the audio signal received by the hearing aid system and the determined geographic location of the hearing aid system; and
A first to the first processor to select a first value of the set of signal processing parameters Θ based on the category of the speech environment determined by the first speech environment detector. Like providing the output,
A first audio environment detector configured;
(D) a user interface to allow a user of the hearing aid system to make adjustments to at least one signal processing parameter θ∈Θ;
(E) a non-transitory medium for recording the adjustment of the at least one signal processing parameter θ∈Θ made by the user of the hearing aid system;
A hearing aid system, wherein the first audio environment detector is configured to provide the first output to the first processor based on the adjustment.   The hearing aid system according to claim 1, wherein the providing of the first output to the first processor is based on an incremental preference derivation based on a Bayesian estimate of the adjustment.   The hearing aid system of claim 1, wherein the location detector comprises a GPS receiver.   The first audio environment detector includes the audio signal received by the hearing aid system, the determined geographical location of the hearing aid system, date, time, speed of the hearing aid system, and the GPS receiver; The method of claim 3, wherein the category of the audio environment surrounding the hearing aid system is determined based on at least one parameter selected from the group consisting of signal strengths of signals received by the hearing aid system. Hearing aid system.   The hearing aid system of claim 1, further comprising a non-transitory medium for recording the geographic location determined by the location detector along with the category of the audio environment at the geographic location.   The first audio environment detector takes into account the probability of occurrence for a previously recorded audio environment category that is within a distance threshold from the determined geographic location, and The hearing aid system according to claim 5, configured to determine a category.   The hearing aid system of claim 1, further comprising a non-transitory medium for storing a particular category of the audio environment along with a respective geographic range.   The location detector automatically accesses the user's calendar system to obtain information about the user's location, otherwise the location detector determines the geographic location of the hearing aid system. The hearing aid system of claim 1, wherein the hearing aid system is configured to determine the geographic location of the hearing aid system based on the information regarding the location of the user when the user is unable to do so.   The first audio environment detector is configured to automatically switch the first hearing aid of the hearing aid system to flight mode when the location detector detects that the user is in an airplane. The hearing aid system according to claim 1.   The first hearing aid comprises at least one posture sensor configured to provide information regarding the posture of the user's head when the user wears the first hearing aid in an intended operating position; The hearing aid system according to claim 1, wherein the first hearing aid is configured to select the first value based on the information regarding the posture of the head of the user.   The hearing aid system according to claim 1, wherein the location detector is part of the first hearing aid.   The hearing aid system of claim 1, further comprising a handheld device communicatively coupled to the first hearing aid and containing the location detector.   The hearing aid system according to claim 12, wherein the handheld device also houses the first audio environment detector.   The hearing aid system of claim 12, wherein the handheld device comprises the user interface.   The hearing aid system according to claim 1, wherein the first hearing aid houses the first sound environment detector. A second microphone for providing a second audio input signal in response to an audio signal received by the second microphone;
A second processor configured to process the second audio input signal according to a signal processing algorithm F (Θ) to generate a second hearing loss compensated audio signal;
A second hearing aid having a second output transducer for providing a second acoustic output signal based on the second hearing loss compensated audio signal;
The first audio environment detector is configured to make a determination of the category of the audio environment based on the first and second audio input signals and the geographic location of the hearing aid system. ,
The hearing aid system according to claim 1.   17. A hearing aid according to claim 16, wherein the first sound environment detector is configured to provide a second output for selecting a second value of the set of signal processing parameters. system. The second hearing aid comprises:
A second audio environment detector,
Making a determination of the category of the audio environment based on the first and second audio input signals and the geographical location of the hearing aid system; and
Providing a second output to the second processor to select a second value of the set of signal processing parameters Θ based on the category determined by the second audio environment detector To do
17. A hearing aid system according to claim 16, comprising a configured second audio environment detector.

Images